Introducing Forezai · TradingAgents — a committee of LLMs decides paper-trades
Up next
Author

📊 Full opportunity report: Introducing Forezai · TradingAgents — a committee of LLMs decides paper-trades on ThorstenMeyerAI.com — validation score, market gap, and execution plan.

TL;DR

Forezai has launched TradingAgents, a system that uses a committee of specialized large language models to generate paper-trades. This development aims to explore AI decision-making in trading without risking real money, marking a step forward in AI research for finance.

Forezai has launched TradingAgents, a system where a committee of large language models (LLMs) collaboratively makes paper-trades based on structured analysis. This development aims to test whether AI-driven multi-agent decision-making can outperform random choices in simulated markets, marking a significant step in AI research for trading strategies.

The new project, Forezai · TradingAgents, is a fork of an open-source multi-agent framework initially designed by TauricResearch. It incorporates operational features such as automated scheduling, paper trading, position management, and multi-broker support, enabling researchers to test AI decision-making in a controlled environment.

Unlike previous experiments with parametric strategies, which largely failed to survive real-market conditions, TradingAgents leverages a structured committee of specialized LLMs that analyze market data through distinct roles—analysts, debate agents, risk teams, and portfolio managers—forcing explicit reasoning and debate among the models. The system does not predict markets but aims to produce consistent, reasoned decisions from the AI committee.

Forezai’s implementation includes safeguards to prevent accidental real-money trading, with multiple layers of refusal to execute live trades. The dashboard provides detailed analytics on model performance, decision rationale, and risk metrics, all running locally without cloud data transmission. The project is designed to serve as a research instrument rather than a commercial trading tool.

Introducing Forezai · TradingAgents — Thorsten Meyer AI
AGENTS
● ANNOUNCEMENT / MAY 2026
THORSTEN MEYER AI · FOREZAI · § 03
FOREZAI · 03
TRADINGAGENTS · LAUNCH
Research Series · Companion to Polybot Week 1-2 · 2026-05-17

Introducing Forezai · TradingAgents.
A committee of LLMs
decides paper-trades.

After two weeks of finding out most parametric strategies don’t work, the obvious next research question: can multi-agent LLM judgment do any better?
A fork of the open-source TradingAgents framework (TauricResearch): thirteen LLM agents in four stages — four parallel analysts · a bull-bear debate with research-manager arbitration · a three-voice risk team · a two-layer trader + portfolio-manager decision. The fork keeps the agent graph intact and adds the operational layer the upstream doesn’t ship: an autonomous loop · a multi-broker abstraction · a local web dashboard · Codex OAuth · MCP plug-ins · 520+ unit tests. The question is narrower than “do LLMs predict the market” — that prior is “no, with high confidence.” The narrower question is: when LLMs are structured into specialised adversarial roles, does the committee produce decisions at least no worse than a coin flip after fees? Honest priors before running: it might fail too. If it appears to work, the most likely explanation is variance.
Thorsten Meyer AI Forezai · TradingAgents Apache-2.0 fork · upstream cited Companion piece · ~2,500 words Polybot · § 03
This is not financial advice. Nothing in this announcement should be used to inform real trading decisions. The software described trades simulated money by default. If you reconfigure it to trade real money, you should expect to lose that money — regardless of how clever any individual agent’s reasoning looks. Algorithmic trading is zero-sum after fees and structurally hostile to part-time retail strategies.
13 agents
Specialised roles in four stages
Analysts · Debate · Risk · Decision
78% / -33%
Polybot prior: fleet win rate
combined with -33% bankroll
520+
Passing unit tests across engine,
services, HTTP routes (starting baseline)
€0 floor
LLM cost on Codex OAuth
(falls back to public API per token)
FOREZAI / TRADINGAGENTS· APACHE 2.0 FORK· UPSTREAM TAURIC RESEARCH· LANGGRAPH· 13 AGENTS / 4 STAGES· 4 PARALLEL ANALYSTS· BULL-BEAR DEBATE· 3-VOICE RISK TEAM· TRADER + PORTFOLIO MANAGER· 5-TIER FINAL RATING· ALPACA PAPER + LOCAL + SHADOW· LIVE ENDPOINTS HARD-REFUSED· FASTAPI + REACT VIA CDN· CODEX OAUTH· MCP PLUG-IN REGISTRY· 520+ UNIT TESTS· POLYBOT WEEK 1: 21 EXPERIMENTS· WEEK 2: -33% BANKROLL· 78% FLEET WIN RATE· HONEST RESEARCH, NOT EDGE· FOREZAI / TRADINGAGENTS· APACHE 2.0 FORK· UPSTREAM TAURIC RESEARCH· LANGGRAPH· 13 AGENTS / 4 STAGES· 4 PARALLEL ANALYSTS· BULL-BEAR DEBATE· 3-VOICE RISK TEAM· TRADER + PORTFOLIO MANAGER· 5-TIER FINAL RATING· ALPACA PAPER + LOCAL + SHADOW· LIVE ENDPOINTS HARD-REFUSED· FASTAPI + REACT VIA CDN· CODEX OAUTH· MCP PLUG-IN REGISTRY· 520+ UNIT TESTS· POLYBOT WEEK 1: 21 EXPERIMENTS· WEEK 2: -33% BANKROLL· 78% FLEET WIN RATE· HONEST RESEARCH, NOT EDGE·
FIG. 01 — THE 13-AGENT COMMITTEE
Thirteen specialised roles · four stages · biases made to argue in public
The architecture forces the system to articulate its reasoning rather than relying on what a single context window happens to recall
Stage 1 · Four analysts in parallel4 agents
Market
Structure, ranges, regime indicators
News + Insider
News flow, filings, insider activity
Fundamentals
Balance sheet, earnings, ratios
Social Sentiment
Social-media tone, retail signal
↓
Stage 2 · Bull-bear debate + research-manager arbitration3 agents
Bull researcher
Argues upside thesis from analyst reports
Bear researcher
Argues downside thesis from same reports
Research manager
Arbitrates · writes single synthesis
↓
Stage 3 · Three-voice risk team3 agents
Aggressive
Looks for upside · accepts variance
Conservative
Looks for downside · protects capital
Neutral
Balances · forces downside articulation
↓
Stage 4 · Two-layer decision2 agents
Trader
Three-tier proposal · buy / hold / sell
Portfolio manager
Five-tier rating + price target + horizon · sees arguments only, never raw data
The portfolio manager only sees the arguments, never the raw data — which forces the committee to make its reasoning explicit rather than relying on a single context window’s recall. The upstream framework ships the agent graph; it does not ship the operational machinery to run that graph on autopilot, observe its results honestly, store them for later inspection, or prevent the operator from accidentally trading real money. That gap is what the Forezai fork fills.
FIG. 02 — THE POLYBOT PRIOR · WHY THIS IS A DIFFERENT BET
Two weeks of paper-trading prediction markets · the trap underneath the headline numbers
25 experiments · 78% fleet-wide win rate · -33% bankroll · most parametric strategies are structurally negative-expectation when measured honestly
The flattering number
78%
Fleet-wide win rate · week 2
“You can win four out of five trades and still go broke, because the one loss is bigger than the four wins put together.” Win rate without P&L context is a mechanical illusion.
The honest number
−33%
Fleet bankroll · week 2 close
The strongest possible demonstration of the trap. A parametric trading strategy that looks compelling in a backtest will almost always fail to survive a fresh sample. Most “edges” are mechanical artefacts.
Week 1: 21 parallel strategy experiments · early winners mostly mechanical illusions · exactly one strategy (a fair-value taker on BTC) showed the mathematical signature of real edge over a few hundred settled trades. Week 2: same fair-value strategy with more data collapsed. A separate mid-week hypothesis (market-making) also failed cleanly. Fleet ended week 2 at roughly negative thirty-three percent of bankroll. The honest research finding wasn’t on the winning side — it was on the losing side. Adding more parameters to Polybot wouldn’t change that. TradingAgents is asking a separable question.
FIG. 03 — WHAT THE FORK ADDS · THE OPERATIONAL LAYER
Six layers the upstream framework doesn’t ship
Same agent graph, intact. The fork makes it a research instrument rather than a tech demo.
01 · Loop
An autonomous loop
Scheduler · watchlist · auto-trader maps ratings to paper orders · allow-list filtering · per-ticker cooldowns · sector caps · cash checks · position manager evaluates open positions every 60s for TP / SL / max-hold. Append-only audit logs.
02 · Brokers
Multi-broker abstraction
Three modes: local Python broker (yfinance fills, JSON-persisted) · Alpaca paper-trading adapter · “shadow” mode running both in parallel with divergence view. Real Alpaca live endpoints are hard-refused at multiple layers.
03 · Dashboard
A local web dashboard
FastAPI backend · React via CDN, no Node toolchain · SVG equity curve · rolling-peak drawdown · win-rate by rating / ticker / model · exit-reason breakdown · LLM cost vs realised P&L joined by run ID. Runs locally; nothing sent to a cloud service.
04 · Codex
Codex OAuth
Runs the engine on a ChatGPT Pro subscription via the Codex backend. LLM cost floor effectively zero if you already have ChatGPT Pro. Token stored encrypted locally. Falls back to the regular OpenAI API if you’d rather pay per token.
05 · Alerts
Multi-channel alerts
Slack · Discord · SMTP email · configurable filter on rating events and order fills · append-only history kept locally. Webhook URLs masked in API responses so a screenshot can’t accidentally leak credentials.
06 · MCP
MCP plug-ins
Registry for adding Anthropic Model Context Protocol servers (Kensho · Aiera · FactSet · Morningstar · LSEG) as analyst tools. Plug-ins advertise category (fundamentals · news · market data · social) · probe endpoint tests credentials.
Honest-by-design touches: every generated report prepends “Research, not advice” and appends a footer with version, commit, provider, models used, run ID, and cost. Closed trades carry the same metadata. 520+ passing unit tests across engine, services, and HTTP routes. The intent: when the system loses money, the journal makes it impossible to pretend it didn’t.
FIG. 04 — HONEST PRIORS · BEFORE RUNNING THIS IN ANGER
Three priors stated before the data starts arriving
The bias of the project: when the data says no, the dashboard says no, the article says no
1
It might fail too. LLMs are not oracles, and a sophisticated framework around language-model outputs does not change the underlying error rate of the model. Sample is still everything. The framework’s outputs are subject to the same statistical noise as any prediction system over small samples.
Highest likelihood
2
If it appears to work, the most likely explanation is variance. The same trap that caught the first article’s candidate edge applies here. A high win rate over fifty trades means much less than it looks. Without out-of-sample confirmation, a flattering early sample tells you almost nothing about whether the system has real edge.
Second-most likely
3
If it appears to work for the right reasons — empirical win rate matches stated confidence, and alpha-versus-benchmark persists across non-overlapping samples — that would be a meaningful research finding. Whether that happens, I don’t know. The point of putting it in the open is that the data will say.
Genuinely open
This is explicitly not a launch announcement for a product anyone should connect a real brokerage account to. The Alpaca live endpoints are hard-refused at multiple layers in the code, and the design choice is deliberate. The right next step is data, not deployment. The bias of the whole project is straightforward: when the data says no, the dashboard says no, the article says no, and no one tries to retroactively rescue the thesis. That’s the contribution.
FIG. 05 — WEEK THREE · WHAT THE METHODOLOGY WILL MEASURE
Four concrete measurements before publishing findings
The hope: write the week-three article from a position of “here’s what the data says”. The fear: another candidate falsified at higher sample. Both outcomes are publishable.
M1 · Sample discipline
Small watchlist for a few weeks before publishing
A handful of tickers across two or three sectors. Long enough to gather sample, narrow enough to keep attention on what’s actually happening per agent. Avoid the noise of a 65-ticker autonomous loop until the smaller version has been read carefully.
M2 · Calibration view
Stated confidence vs. realised win rate
When the system says “75% confident”, do the trades actually win 75% of the time? Same measurement applied to Polybot’s fair-value model. If the model is systematically over-confident, that bias dominates everything downstream.
M3 · Cost accounting
Cost per ticker · per rating · per profitable trade
With Codex OAuth the marginal LLM cost is effectively zero. With the public OpenAI API, each run is hundreds of agent turns. The honest question: does this scale economically if you ever did run it at real cost?
M4 · Non-overlapping windows
Alpha vs benchmark · out-of-sample
Not within-sample alpha — trivially inflatable. Hold out one period entirely, run the system on the next, then check whether the held-out result matches the in-sample stats. If they diverge sharply, the in-sample was curve-fit.
Open under Apache-2.0 with upstream cited from every relevant surface. Not open: the operator’s running results, the specific watchlist, the per-agent prompt customisations, the alert channels, the trade journals — kept local for the same reason Polybot’s per-experiment data is kept local. Publishing exact configurations encourages people to copy them with real money, which is the opposite of what an honest research project should do. Summary findings will be published. Recipes will not.
The bet is on a different mechanism, not a different parameter setting. The point is not to find a money-printing AI. The point is to put honest measurements of these systems into the public record — so the next person looking at the space starts a step further along than the last.
Thorsten Meyer AI · Introducing Forezai · TradingAgents · § 03
Source dossier & project notes
  • Forezai · TradingAgents fork — Apache-2.0 license preserved · GitHub published under same author as Polybot · upstream cited from README · marketing pages · generated report footers · forezai.com
  • TradingAgents upstream — TauricResearch team · multi-agent stock-research framework on LangGraph · the agent graph is intact in the fork · upstream license requirements (notice file · attribution · patent-grant clauses) preserved if anyone forks again
  • Polybot · Week 1 — 21 parallel strategy experiments on Polymarket 5-minute Up/Down · early winners mostly mechanical illusions · one BTC fair-value taker showed mathematical signature of edge over a few hundred settled trades
  • Polybot · Week 2 — fair-value strategy collapsed at higher sample · market-making hypothesis failed cleanly · fleet ended at ~negative 33% of bankroll · 78% fleet-wide win rate combined with deeply negative P&L · “you can win four out of five trades and still go broke”
  • Architecture · Stage 1 — Four analysts in parallel: market structure · news + insider · fundamentals (balance sheet + earnings) · social-media sentiment · each writes a short structured report independently
  • Architecture · Stage 2 — Bull-bear debate: two researcher agents argue opposing theses from the analyst reports · a research-manager agent arbitrates and writes a single synthesis
  • Architecture · Stage 3 — Three-voice risk team: aggressive (upside, accepts variance) · conservative (downside, protects capital) · neutral (balances) · forces explicit articulation of downside before any order is proposed
  • Architecture · Stage 4 — Two-layer decision: trader agent produces three-tier proposal (buy / hold / sell) · portfolio-manager agent synthesises into final five-tier rating with price target and time horizon · PM sees arguments only, never raw data
  • Operational layer — Autonomous loop · multi-broker abstraction (local Python / Alpaca paper / shadow mode) · FastAPI + React-via-CDN web dashboard · Codex OAuth on ChatGPT Pro · multi-channel alerts (Slack · Discord · SMTP) · MCP plug-in registry (Kensho · Aiera · FactSet · Morningstar · LSEG)
  • Engineering baseline — 520+ passing unit tests across engine, services, HTTP routes · honest-by-design touches: “Research, not advice” prepended to every report · footer with version, commit, provider, models used, run ID, cost
  • Hard-refused live trading — Real Alpaca live endpoints hard-refused at multiple layers · operator must deliberately override the refusal in more than one place to actually risk real money
  • Week three methodology — Small watchlist · calibration view (stated confidence vs realised win rate) · cost accounting per ticker / rating / profitable trade · out-of-sample alpha across non-overlapping windows
  • Stance — Summary findings published · recipes kept local · the bias is straightforward: when the data says no, the dashboard says no, the article says no, no one tries to retroactively rescue the thesis
Colophon

Set in Source Serif 4 (display, italic accent), EB Garamond (body), IBM Plex Sans (UI labels), IBM Plex Mono (mastheads, ticker, stamps, plug-in tags). Paper-cool gray-cream #e6e7e4.

Chromatic register: structural-slate dominant (committee-architecture analysis), empirical-clay for the Polybot prior forensic, labor-rose for the cautionary “expect to lose money” stance and the regulatory disclaimers, transition-bronze for the week-three forward-shape methodology, alternative-sage for the open-source / honest-measurement positive signal.

Key frame:

FOREZAI / TRADINGAGENTS 13-AGENT COMMITTEE POLYBOT PRIOR OPERATIONAL LAYER EXPECT TO LOSE MONEY CALIBRATION VIEW BULL-BEAR DEBATE HARD-REFUSED LIVE APACHE 2.0 HONEST MEASUREMENTS

Potential Impact of AI Multi-Agent Decision Systems

This development is significant because it tests whether structured AI committees can generate more reliable trading decisions than individual models or simple algorithms. If successful, it could influence future AI research and development in financial decision-making, emphasizing explicit reasoning and debate among models rather than raw prediction accuracy.

While the system currently operates in simulation, it provides a foundation for understanding how AI can be used to simulate complex decision processes, which may eventually translate into more robust automated trading systems or decision-support tools for human traders.

Chemical Process Simulation and the Aspen HYSYS Software

Chemical Process Simulation and the Aspen HYSYS Software

Used Book in Good Condition

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background on AI in Trading and Recent Research Findings

Previous experiments with parametric trading strategies, such as those tested by Thorsten Meyer’s team, showed that many seemingly effective rules tend to fail under real-market conditions, often losing money despite high win rates. These findings highlighted the risk of overfitting and the importance of explicit reasoning in trading algorithms.

The question then shifted toward whether less rule-bound AI systems, such as committees of LLMs structured to argue and debate, could produce more reliable decisions. TauricResearch’s TradingAgents framework was developed to explore this, with its multi-agent architecture designed to simulate a team of analysts and decision-makers.

Forezai’s fork enhances this framework by adding operational tools for autonomous, scheduled testing, paper trading, and detailed analytics, making it more suitable for serious research and experimentation.

“This system allows us to test whether a structured committee of LLMs can produce decisions that are at least no worse than random, which is a foundational question in AI-driven trading research.”

— Thorsten Meyer, researcher

Agentic Architectural Patterns for Building Multi-Agent Systems: Proven design patterns and practices for GenAI, agents, RAG, LLMOps, and enterprise-scale AI systems

Agentic Architectural Patterns for Building Multi-Agent Systems: Proven design patterns and practices for GenAI, agents, RAG, LLMOps, and enterprise-scale AI systems

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Unanswered Questions About AI Decision-Making Efficacy

It remains unclear whether the AI committee can consistently outperform simple or random strategies in live or more complex simulated environments. The system currently operates only in paper trading, and its long-term effectiveness and potential to adapt to real market conditions are still untested.

Additionally, the extent to which explicit reasoning among models improves decision quality compared to traditional prediction models is still under investigation.

The No-BS Guide to AI for Trading & Market Research: How to Use ChatGPT, Claude & AI Tools for Market Analysis, Stock Research & Data-Driven Trading ... ... Required (The No-BS AI Playbooks Book 3)

The No-BS Guide to AI for Trading & Market Research: How to Use ChatGPT, Claude & AI Tools for Market Analysis, Stock Research & Data-Driven Trading … … Required (The No-BS AI Playbooks Book 3)

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Next Steps for Testing and Development of AI Trading Agents

Future work will focus on extended testing of the system over longer periods and different market conditions to evaluate robustness. Researchers plan to refine the agent roles, incorporate more complex data sources, and explore real-time decision-making capabilities, always with safeguards to prevent real-money trading until proven reliable.

Further developments may include integrating human oversight, expanding multi-broker support, and publishing detailed performance analytics to assess the system’s potential for practical deployment.

AI ROI Is Not a Dashboard: Why Enterprises Need Evidence, Governance, and Financial Control (Trajecta Research Book 5)

AI ROI Is Not a Dashboard: Why Enterprises Need Evidence, Governance, and Financial Control (Trajecta Research Book 5)

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Can Forezai TradingAgents be used for live trading?

No, currently the system is designed for paper trading and research purposes only. It includes safeguards to prevent accidental live trading.

How does the AI committee make decisions?

The system uses multiple specialized LLMs that analyze market data through different perspectives, debate, and synthesize their findings into a final recommendation, with explicit reasoning articulated at each step.

What advantages does a multi-agent system have over single models?

It encourages explicit reasoning, debate, and diverse perspectives, which may lead to more reliable and robust decision-making compared to individual models or rule-based systems.

When will this system be tested in live markets?

There are no current plans for live trading; future steps involve extended simulation testing and validation before considering real-market deployment.

Is this system intended for commercial use?

No, Forezai TradingAgents is primarily a research platform aimed at understanding AI decision-making processes in trading, not a commercial trading product.

Source: ThorstenMeyerAI.com

This content is for general information only and is not financial, tax or legal advice. Consult a qualified professional for decisions about your money.
You May Also Like
scammers increasingly target victims

Pig Butchering Scams Have Surged 40%, as Scammers Ramp up Both Speed and Severity.

Pig butchering scams are on the rise, increasing by 40%, leaving many wondering how to protect themselves from these deceptive tactics.
crypto market future predictions

Crypto Market Outlook 2026: Key Themes and Expectations

Discover the transformative trends shaping the crypto market by 2026, as new opportunities arise and expectations shift dramatically. What will the future hold?
The Bubble Is Not in Valuations: It’s in the Productivity Gap

The Bubble Is Not in Valuations: It’s in the Productivity Gap

Analysis of the disconnect between AI valuation premiums and actual productivity gains, highlighting the risk of a structural bubble in expectations.
The Compute Reckoning: Anthropic Finally Admits What Customers Suspected for Ten Months

The Compute Reckoning: Anthropic Finally Admits What Customers Suspected for Ten Months

Anthropic reveals that its recent customer experience issues were due to compute shortages, after years of suspected capacity constraints.